Identification of Individuals at Increased Risk for Pancreatic Cancer in a Community-Based Cohort of Patients With Suspected Chronic Pancreatitis. Clin Transl Gastroenterol. 2020;11:e00147. [PMID: 32352677 PMCID: PMC7263650 DOI: 10.14309/ctg.0000000000000147]

Pancreatic Cancer: A Review of Risk Factors

Pancreatic adenocarcinoma is one of the most lethal types of gastrointestinal cancer despite the latest medical advances. Its incidence has continuously increased in recent years in developed countries. The location of the pancreas can result in the initial symptoms of neoplasia being overlooked, which can lead to a delayed diagnosis and a subsequent reduction in the spectrum of available therapeutic options. The role of modifiable risk factors in pancreatic cancer has been extensively studied in recent years, with smoking and alcohol consumption identified as key contributors. However, the few screening programs that have been developed focus exclusively on genetic factors, without considering the potential impact of modifiable factors on disease occurrence. Thus, fully understanding and detecting the risk factors for pancreatic cancer represents an important step in the prevention and early diagnosis of this type of neoplasia. This review reports the available evidence on different risk factors and identifies the areas that could benefit the most from additional studies.

Machine Learning Models for Pancreatic Cancer Risk Prediction Using Electronic Health Record Data-A Systematic Review and Assessment

INTRODUCTION

Accurate risk prediction can facilitate screening and early detection of pancreatic cancer (PC). We conducted a systematic review to critically evaluate effectiveness of machine learning (ML) and artificial intelligence (AI) techniques applied to electronic health records (EHR) for PC risk prediction.

METHODS

Ovid MEDLINE(R), Ovid EMBASE, Ovid Cochrane Central Register of Controlled Trials, Ovid Cochrane Database of Systematic Reviews, Scopus, and Web of Science were searched for articles that utilized ML/AI techniques to predict PC, published between January 1, 2012, and February 1, 2024. Study selection and data extraction were conducted by 2 independent reviewers. Critical appraisal and data extraction were performed using the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies checklist. Risk of bias and applicability were examined using prediction model risk of bias assessment tool.

RESULTS

Thirty studies including 169,149 PC cases were identified. Logistic regression was the most frequent modeling method. Twenty studies utilized a curated set of known PC risk predictors or those identified by clinical experts. ML model discrimination performance (C-index) ranged from 0.57 to 1.0. Missing data were underreported, and most studies did not implement explainable-AI techniques or report exclusion time intervals.

DISCUSSION

AI/ML models for PC risk prediction using known risk factors perform reasonably well and may have near-term applications in identifying cohorts for targeted PC screening if validated in real-world data sets. The combined use of structured and unstructured EHR data using emerging AI models while incorporating explainable-AI techniques has the potential to identify novel PC risk factors, and this approach merits further study.

Identification of pancreatic cancer risk factors from clinical notes using natural language processing

OBJECTIVES

Screening for pancreatic ductal adenocarcinoma (PDAC) is considered in high-risk individuals (HRIs) with established PDAC risk factors, such as family history and germline mutations in PDAC susceptibility genes. Accurate assessment of risk factor status is provider knowledge-dependent and requires extensive manual chart review by experts. Natural Language Processing (NLP) has shown promise in automated data extraction from the electronic health record (EHR). We aimed to use NLP for automated extraction of PDAC risk factors from unstructured clinical notes in the EHR.

METHODS

We first developed rule-based NLP algorithms to extract PDAC risk factors at the document-level, using an annotated corpus of 2091 clinical notes. Next, we further improved the NLP algorithms using a cohort of 1138 patients through patient-level training, validation, and testing, with comparison against a pre-specified reference standard. To minimize false-negative results we prioritized algorithm recall.

RESULTS

In the test set (n = 807), the NLP algorithms achieved a recall of 0.933, precision of 0.790, and F1-score of 0.856 for family history of PDAC. For germline genetic mutations, the algorithm had a high recall of 0.851, while precision and F1-score were lower at 0.350 and 0.496 respectively. Most false positives for germline mutations resulted from erroneous recognition of tissue mutations.

CONCLUSIONS

Rule-based NLP algorithms applied to unstructured clinical notes are highly sensitive for automated identification of PDAC risk factors. Further validation in a large primary-care patient population is warranted to assess real-world utility in identifying HRIs for pancreatic cancer screening.

Diagnostic Differentiation between Pancreatitis and Pancreatic Cancer: A Scoping Review

Pancreatitis, encompassing acute and chronic forms, and pancreatic cancer pose significant challenges to the exocrine tissue of the pancreas. Recurrence rates and complications following acute pancreatitis episodes can lead to long-term risks, including diabetes mellitus. Chronic pancreatitis can develop in approximately 15% of cases, regardless of the initial episode's severity. Alcohol-induced pancreatitis, idiopathic causes, cigarette smoking, and hereditary pancreatitis contribute to the progression to chronic pancreatitis. Chronic pancreatitis is associated with an increased risk of pancreatic cancer, with older age at onset and smoking identified as risk factors. This scoping review aims to synthesise recent publications (2017-2022) on the diagnostic differentiation between pancreatitis and pancreatic cancer while identifying knowledge gaps in the field. The review focuses on biomarkers and imaging techniques in individuals with pancreatitis and pancreatic cancer. Promising biomarkers such as faecal elastase-1 and specific chemokines offer non-invasive ways to assess pancreatic insufficiency and detect early biomarkers for chronic pancreatitis. Imaging techniques, including computed tomography (CT), magnetic resonance imaging (MRI), endoscopic ultrasound (EUS), and positron emission tomography (PET), aid in differentiating between chronic pancreatitis and pancreatic cancer. However, accurately distinguishing between the two conditions remains a challenge, particularly when a mass is present in the head of the pancreas. Several knowledge gaps persist despite advancements in understanding the association between pancreatitis and pancreatic cancer, including the correlation between histopathological grading systems, non-invasive imaging techniques, and biomarkers in chronic pancreatitis to determine the risk of progression to pancreatic cancer, as well as differentiating between the two conditions. Further research is necessary to enhance our understanding of these aspects, which can ultimately improve the diagnosis and management of pancreatitis and pancreatic cancer.

The Impact of Chronic Pancreatitis on the Occurrences of Human Cancers: Real-World Data

Chronic pancreatitis (CP) may induce systemic inflammation, potentially increasing cancer susceptibility. However, the link between CP and extra-pancreatic cancer remains underexplored. Employing Taiwanese National Health Insurance Database data from 2000 to 2017, we compared 5394 CP patients with 21,576 non-CP individuals through propensity score matching. CP patients exhibited a significantly higher cancer risk (adjusted hazard ratio (aHR) of 1.32 for females and 1.68 for males) and cumulative incidence (p < 0.001) compared to non-CP individuals. CP showed notable associations with pancreatic (aHR = 3.51), liver (aHR = 1.62), stomach (aHR = 2.01), and other cancers (aHR = 2.09). In terms of liver cancer, CP was significantly associated with patients without viral hepatitis, regardless of gender (aHR = 2.01 for women; aHR = 1.54 for men). No significant cancer occurrences were observed within the first year following CP diagnosis. Pancreatic or liver cancer developed in approximately half of CP patients within 2-3 years, while gastric cancer in male CP patients predominantly occurred around the fifth year after diagnosis. These findings inform potential cancer-screening plans for CP patients.

Recommended practices and ethical considerations for natural language processing-assisted observational research: A scoping review

An increasing number of studies have reported using natural language processing (NLP) to assist observational research by extracting clinical information from electronic health records (EHRs). Currently, no standardized reporting guidelines for NLP-assisted observational studies exist. The absence of detailed reporting guidelines may create ambiguity in the use of NLP-derived content, knowledge gaps in the current research reporting practices, and reproducibility challenges. To address these issues, we conducted a scoping review of NLP-assisted observational clinical studies and examined their reporting practices, focusing on NLP methodology and evaluation. Through our investigation, we discovered a high variation regarding the reporting practices, such as inconsistent use of references for measurement studies, variation in the reporting location (reference, appendix, and manuscript), and different granularity of NLP methodology and evaluation details. To promote the wide adoption and utilization of NLP solutions in clinical research, we outline several perspectives that align with the six principles released by the World Health Organization (WHO) that guide the ethical use of artificial intelligence for health.

Incidence and risk of pancreatic cancer in patients with chronic pancreatitis: defining the optimal subgroup for surveillance

We aimed to present the incidence and risk factors for pancreatic cancer in a multicenter retrospective cohort of patients with chronic pancreatitis (CP). Patients with ICD-10 codes for CP (K86.0, K86.1) who underwent abdominal CT or MRI between January 2010 and December 2021 in seven academic hospitals were analyzed. After exclusions, we identified 727 patients with definite CP with a median follow-up of 3.6 years (range 1.0‒12.9). During 3290 person-years of observation, pancreatic cancers were diagnosed in 16 patients (2.20%, 0.49% per year) after a median follow-up of 2.4 years (range 1.4‒6.6), with an age- and sex-standardized incidence ratio of 18.1 (95% CI 10.4‒29.5). The underlying CPs in the 16 pancreatic cancers were classified as chronic obstructive pancreatitis (10, 63%), chronic obstructive and calcifying pancreatitis (4, 25%), chronic calcifying pancreatitis (1, 6%), and autoimmune pancreatitis (1, 6%). Factors associated with pancreatic cancer development included age (HR 4.830, p = 0.006), parenchymal calcification (HR 0.213, p = 0.003), pancreatic duct stricture (HR 2.706, p = 0.048), and serum CA 19‒9 level (HR 3.567, p = 0.014). After adjustment, age over 60 years (HR 4.540, p = 0.009) and serum CA 19‒9 levels greater than 100 U/mL (HR 3.528, p = 0.015) were independent risk factors for pancreatic cancer.

Pancreatitis and Pancreatic Cancer Risk

Purpose: The present retrospective study aimed to explore the relationship between pancreatitis and pancreatic cancer in the population cohort of the UK Biobank (UKB) (https://www.ukbiobank.ac.uk). Methods: From the 500 thousand population cohort of UKB, according to the age and gender of patients with pancreatic cancer 1:10, matching the control without pancreatic cancer, the binary Logistic regression model was used to analyze the relationship between pancreatitis and pancreatic cancer, and subgroup analyses were used to identify potential effect modifiers. Results: A total of 1538 patients with pancreatic cancer were compared with 15 380 controls. In the fully adjusted model, patients with pancreatitis had a significantly increased risk of pancreatic cancer compared with no pancreatitis. The risk of pancreatitis and pancreatic cancer increased with the age of pancreatitis, and the risk of pancreatic cancer was highest in the 61 to 70 age group. In addition, in the first 3 years of acute pancreatitis, the risk of pancreatic cancer increased significantly with the increase in the duration of the disease (odds ratio [OR] 29.13, 95% confidence interval [CI]: 16.34-51.93), after 3 years, the trend of increase decreased. After more than 10 years, there was no significant correlation between the risk of acute pancreatitis and pancreatic cancer. However, patients with chronic pancreatitis were significantly associated with an increased risk of pancreatic cancer only in the first 3 years (OR 28.14, 95% CI: 14.86-53.31). Conclusion: Pancreatitis may associate with an increased risk of pancreatic cancer. The older the age of pancreatitis, the higher the risk of pancreatic cancer. The risk of pancreatic cancer increases significantly in the first 3 years of the course of pancreatitis. This may provide an alternative strategy for the early identification of individuals at high risk of pancreatic cancer.

Clinical and Imaging Predictors of Pancreatic Cancer in Patients Hospitalized for Acute Pancreatitis

Background and Aims

Identifying factors associated with increased short-term risk of pancreatic cancer in the setting of acute pancreatitis (AP) can inform clinical care decisions and expedite cancer diagnosis.

Methods

A retrospective cohort study of patients hospitalized for AP between 2007 and 2017 in an integrated health-care system in Southern California. AP cases were identified by diagnosis code with laboratory confirmation. Multivariable Cox proportional hazards regression model was used to assess risk of pancreatic cancer within 3 years of AP, adjusting for patient demographics, clinical parameters (body mass index, AP etiology, chronic pancreatitis, diabetes) and radiographic imaging features.

Results

Among 9,490 patients hospitalized with AP, the mean (standard deviation) age was 55.8 (17.8) years, 55% were women, and 42% were Hispanic. Majority of AP cases were biliary (64%), 12% were alcohol-related, 5% were hypertriglyceridemia-induced, and 19% were other/unknown etiology. Ninety-five (1%) patients were diagnosed with pancreatic cancer within 3 years of AP (4.2 cases/1000 person-years). Risk factors for pancreatic cancer were age ≥65 years (hazard risk [HR]: 2.5, 95% confidence interval [CI]: 1.2-5.3), male sex (HR: 1.9, 95% CI: 1.2-2.8), Asian/Pacific Islander race (HR: 2.0, 95% CI: 1.1-3.6), and underweight body mass index (HR: 2.6, 95% CI: 1.1-6.5). In addition, other/unknown AP etiology (HR: 2.0, 95% CI: 1.3-3.1) and dilatation of the main pancreatic duct (HR: 6.6, 95% CI: 4.2-10.5) were independently associated with increased risk of pancreatic cancer.

Conclusion

In addition to older age, the lack of well-established etiology, underweight body habitus, and main pancreatic duct dilatation were independently associated with increased short-term risk of pancreatic cancer among patients hospitalized for AP.

Risk Prediction of Pancreatic Cancer in Patients With Abnormal Morphologic Findings Related to Chronic Pancreatitis: A Machine Learning Approach

BACKGROUND AND AIMS

A significant factor contributing to poor survival in pancreatic cancer is the often late stage at diagnosis. We sought to develop and validate a risk prediction model to facilitate the distinction between chronic pancreatitis-related vs potential early pancreatic ductal adenocarcinoma (PDAC)-associated changes on pancreatic imaging.

METHODS

In this retrospective cohort study, patients aged 18-84 years whose abdominal computed tomography/magnetic resonance imaging reports indicated duct dilatation, atrophy, calcification, cyst, or pseudocyst between January 2008 and November 2019 were identified. The outcome of interest is PDAC in 3 years. More than 100 potential predictors were extracted. Random survival forests approach was used to develop and validate risk models. Multivariable Cox proportional hazard model was applied to estimate the effect of the covariates on the risk of PDAC.

RESULTS

The cohort consisted of 46,041 (mean age 66.4 years). The 3-year incidence rate was 4.0 (95% confidence interval CI 3.6-4.4)/1000 person-years of follow-up. The final models containing age, weight change, duct dilatation, and either alkaline phosphatase or total bilirubin had good discrimination and calibration (c-indices 0.81). Patients with pancreas duct dilatation and at least another morphological feature in the absence of calcification had the highest risk (adjusted hazard ratio [aHR] = 14.15, 95% CI 8.7-22.6), followed by patients with calcification and duct dilatation (aHR = 7.28, 95% CI 4.09-12.96), and patients with duct dilation only (aHR = 6.22, 95% CI 3.86-10.03), compared with patients with calcifications alone as the reference group.

CONCLUSION

The study characterized the risk of pancreatic cancer among patients with 5 abnormal morphologic findings based on radiology reports and demonstrated the ability of prediction algorithms to provide improved risk stratification of pancreatic cancer in these patients.

Chronic Pancreatitis Is a Risk Factor for Pancreatic Cancer, and Incidence Increases With Duration of Disease: A Systematic Review and Meta-analysis

INTRODUCTION

Observational studies have suggested an increased risk of pancreatic ductal adenocarcinoma (PDAC) in patients with acute and chronic pancreatitis. We conducted a systematic review and meta-analysis to evaluate the magnitude of this association and summarize the published epidemiological evidence.

METHODS

We searched electronic databases (MEDLINE, Embase, Web of Science, Cochrane, and Scopus) and reference lists until January 18, 2021. Studies reporting quantitative association between pancreatitis and PDAC were included and assessed for eligibility, data abstraction, and risk of bias. Standardized incidence ratios (SIRs) were pooled using the random-effects model.

RESULTS

Twenty-five cohort and case-control studies met inclusion criteria. Meta-analysis of 12 chronic pancreatitis (CP) studies demonstrated an increased risk of PDAC in patients with CP (SIR: 22.61, 95% confidence interval [CI]: 14.42-35.44). This elevated risk persisted in subgroup analysis of studies that excluded patients diagnosed with PDAC within 2 years of CP diagnosis (SIR: 21.77, 95% CI: 14.43-32.720). The risk was higher in hereditary pancreatitis (SIR: 63.36, 95% CI: 45.39-88.46). The cumulative incidence rates of PDAC in CP increased with follow-up duration. Limited evidence in acute pancreatitis indicates higher PDAC risk in the subset of patients eventually diagnosed with CP. PDAC seems to be uncommon in patients with autoimmune pancreatitis, with 8 reported cases in 358 patients with autoimmune pancreatitis across 4 studies.

DISCUSSION

There is an increased risk of PDAC in patients with CP, and incidence rates increase with CP disease duration. Our results indicate that PDAC surveillance may be considered in individuals with long-standing CP.

Pancreatitis and pancreatic cancer: A case of the chicken or the egg

Acute pancreatitis (AP), chronic pancreatitis (CP) and pancreatic cancer are three distinct pancreatic diseases with different prognoses and treatment options. However, it may be difficult to differentiate between benign and malignant disease. AP may be a first symptom of pancreatic cancer, particularly in patients between the ages of 56 and 75 with presumed idiopathic AP who had a concomitant diagnosis of new-onset diabetes mellitus or patients who present with CP at diagnosis of AP. In these patients, additional imaging is warranted, preferably by endoscopic ultrasonography. CP may lead to pancreatic cancer through oncogenic mutations, mostly in patients with hereditary CP, and in patients in whom risk factors for pancreatic cancer (e.g., nicotine and alcohol abuse) are also present. Patients with PRSS1-mediated CP and patients with a history of autosomal dominant hereditary CP without known genetic mutations may be considered for surveillance for pancreatic cancer. Pancreatic inflammation may mimic pancreatic cancer by appearing as a focal mass-forming lesion on imaging. Differentiation between the above mentioned benign and malignant disease may be facilitated by specific features like the duct-penetrating sign and the duct-to-parenchyma ratio. Research efforts are aimed towards developing a superior discriminant between pancreatitis and pancreatic cancer in the form of imaging modalities or biomarkers. This may aid clinicians in timely diagnosing pancreatic cancer in a potentially curable stage.

Lifetime smoking history and cohort-based smoking prevalence in chronic pancreatitis

BACKGROUND/OBJECTIVE

Smoking prevalence in patients with chronic pancreatitis [CP] is high. We aimed to understand lifetime history of smoking and cohort trends in CP patients to inform effective strategies for smoking cessation.

METHOD

Data on 317 CP patients from the North American Pancreatitis Study 2 [NAPS2] Continuation and Validation Study and the NAPS2 Ancillary Study were analyzed. Smoking history was assessed for each phase of life from the onset of smoking to study enrollment. Data on second-hand smoke and drinking history were also collected. We compared demographic factors, drinking history, pain level and pancreas morphology by smoking status at age 25 (non-smoking, <1 pack per day [PPD], ≥1 PPD). We compared smoking prevalence by birth cohorts: 1930-1949, 1950-1969, 1970-1989.

RESULT

Fifty-one percent of CP patients reported smoking at the time of enrollment. Those who smoked ≥1 PPD at age 25 smoked a cumulative total of 30.3 pack-years of cigarettes over a lifetime. Smoking at age 25 was associated with greater lifetime drinking and greater exposure to second-hand smoke at home and at workplace. Pancreatic atrophy and pseudocysts were more common among smokers. Pancreatic pain was more severe among smokers, and 12-13% of smokers reported smoking to alleviate pain. Male CP patients born in 1950-1969 reported the highest peak prevalence of smoking, and female CP patients born in 1970-1989 reported highest peak prevalence of smoking.

CONCLUSION

CP patients exhibit intense and sustained smoking behavior once established in the 20s. Regardless, cohort analyses demonstrate that the behaviors could potentially be altered by policy changes.

Artificial Intelligence and Early Detection of Pancreatic Cancer: 2020 Summative Review

ABSTRACT

Despite considerable research efforts, pancreatic cancer is associated with a dire prognosis and a 5-year survival rate of only 10%. Early symptoms of the disease are mostly nonspecific. The premise of improved survival through early detection is that more individuals will benefit from potentially curative treatment. Artificial intelligence (AI) methodology has emerged as a successful tool for risk stratification and identification in general health care. In response to the maturity of AI, Kenner Family Research Fund conducted the 2020 AI and Early Detection of Pancreatic Cancer Virtual Summit (www.pdac-virtualsummit.org) in conjunction with the American Pancreatic Association, with a focus on the potential of AI to advance early detection efforts in this disease. This comprehensive presummit article was prepared based on information provided by each of the interdisciplinary participants on one of the 5 following topics: Progress, Problems, and Prospects for Early Detection; AI and Machine Learning; AI and Pancreatic Cancer-Current Efforts; Collaborative Opportunities; and Moving Forward-Reflections from Government, Industry, and Advocacy. The outcome from the robust Summit conversations, to be presented in a future white paper, indicate that significant progress must be the result of strategic collaboration among investigators and institutions from multidisciplinary backgrounds, supported by committed funders.